Characteristics, treatment patterns, healthcare resource use, and costs among adult patients diagnosed with neurofibromatosis type 1 and plexiform neurofibromas in the United States Authors: Xiaoqin Yang, Mavis Obeng-Kusi, Theresa Dettling, Ayo Adeyemi, Marianne Cunnington, Qing Liu, Dominique Lejeune, Grace Chen, Mei Sheng Duh, Justin T Jordan Published: 2026 ABSTRACT AbstractBackgroundNeurofibromatosis type 1 (NF1) is a progressive rare genetic disorder that frequently involves the development of plexiform neurofibromas (PN). Until recently, there was no approved pharmacological treatment for adults with NF1-PN, and the disease burden is not well understood. This real-world retrospective study aimed to describe treatment patterns, healthcare resource utilization (HRU), and costs among adults with NF1-PN in the US.MethodsData were obtained from the Merative™ MarketScan® Commercial and Medicare Databases (01/01/2016-12/31/2022). Adults diagnosed with NF1 and PN were matched to controls without NF1-PN on sociodemographic characteristics (1:5 ratio). Continuous enrollment for ≥ 12 months before and ≥ 3 months after the index diagnosis (defined as the latter of NF1 or PN diagnosis by ICD-10-CM codes) was required for all patients. Treatment patterns, all-cause, and NF1/PN-specific HRU and costs were characterized during the follow-up period. A generalized linear model with Poisson or Tweedie distribution was used to compare all-cause HRU and costs, respectively, between patients and controls.ResultsThis study included 944 patients with NF1-PN and 4,720 controls. The mean age was 39.6 (standard deviation: 15.6) years and 59.6% were female. Over a mean follow-up of 25.8 months, the most common treatment among patients with NF1-PN was prescription pain medication (69.5%), followed by debulking surgeries (22.9%), cytotoxic chemotherapy (7.0%), radiotherapy (4.8%), and targeted therapies (e.g., MEK inhibitors) (4.7%). All-cause HRU was significantly higher among patients than controls across all settings (1.6 vs. 0.27 inpatient days per patient per year [PPPY], 18.8 vs. 9.1 outpatient visits PPPY, 0.80 vs. 0.53 emergency department visits PPPY), with adjusted incidence rate ratios of 1.4 to 4.2 (all p < 0.001). Among patients with NF1-PN, 53% and 18% of all-cause inpatient days and outpatient visits, respectively, were attributable to an NF1 or PN diagnosis. The adjusted mean total healthcare costs were $23,516 PPPY higher among the NF1-PN ($34,398 PPPY) versus matched control ($6,149 PPPY) cohort, with a cost ratio of 4.3 (p < 0.001).ConclusionsThis real-world study identified a substantially higher HRU and economic burden among adult patients with versus without NF1-PN across all settings, highlighting the need for new treatments to manage NF1-PN among this population.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12883-026-04868-8.Keywords: Neurofibromatosis type 1, Plexiform neurofibromas, Treatment patterns, Healthcare resource use, Healthcare costs, Retrospective matched cohort study Background Neurofibromatosis type 1 (NF1) is a rare, autosomal dominant disorder characterized by tumors of the central and peripheral nervous system, estimated to affect 1 in 3,200 people [1–3]. Patients with NF1 may experience many non-neoplastic consequences of the disease, including cognitive impairment, skeletal abnormalities or vascular disease, as well as manifold neoplastic consequences including cutaneous neurofibromas, malignant peripheral nerve sheath tumors, or plexiform neurofibromas (PNs), the latter of which are present in up to 50% of patients [1, 2, 4, 5]. PNs are complex nerve sheath tumors that may grow progressively throughout a patient’s lifetime and may be associated with pain, neurologic dysfunction, disfigurement, and decreased quality of life [6–8]. The standard of care for PN treatment among patients with NF1 was historically limited to surgical tumor debulking [9–11]. However, complete tumor resection is generally not feasible because of the large size of many PNs, and their adjacency to major nervous system or vascular structures [9, 10]. Surgery is typically postponed for as long as possible due to the involvement of multiple fascicles of nerves and tissues that increase the risk of neurological and functional damage [9, 10]. Instead, pharmaceuticals are quickly becoming a consideration for first-line treatment of NF1-PN [11, 12], including with inhibition of mitogen-activated protein kinase (MEK) 1/2 proteins of the mitogen-activated protein kinases/extracellular signal-regulated kinases (MAPK-ERK) signaling pathway, which becomes dysregulated upon mutation of the NF1 gene [13, 14]. The MAPK-ERK pathway is critical for control of cellular proliferation, and inhibition thereof results in proliferation arrest and apoptosis, reducing tumor growth [13, 14]. Selumetinib, an oral, highly selective, MEK inhibitor, was approved by the United States (US) Food and Drug Administration (FDA) in April 2020 and November 2025 for pediatric and adult patients with NF1, respectively, who have symptomatic and inoperable PN [15–17]. In the Phase 2 SPRINT trial, 68% of pediatric patients with NF1-PN treated with selumetinib experienced a confirmed partial response, defined as a target neurofibroma volume decrease from baseline of ≥ 20% on consecutive restaging examinations ≥ 3 months apart [18]. Subsequent independent central review confirmed 44% partial response rate at the time of FDA approval [19]. Likewise, in the international, randomized, double-blind, placebo-controlled Phase 3 KOMET trial, a significantly greater proportion of adult patients with NF1-PN experienced a reduction in PN tumor volume with selumetinib versus placebo treatment, and the safety profile was similar to that of pediatric patients. [20, 21]. Selumetinib led to a rapid onset of response (median: 3.7 months), with an objective response rate (ORR) of 20% for patients randomized to selumetinib compared with 5% for those randomized to placebo [21]. Additionally, patients with a baseline chronic pain score of ≥ 3 on the 11-point Numerical Rating Scale had a clinically meaningful reduction in pain with selumetinib versus placebo treatment [21]. A higher ORR of 64% was observed in a Phase 2 trial of selumetinib among adult patients with a similar study design [22]. Mirdametinib, another MEK inhibitor, was approved by the FDA in February 2025 for adults and children aged ≥ 2 years with NF1 and associated PNs that are symptomatic and not amenable to complete resection. Based on the single-arm Phase 2 ReNeu study, 41% of adults and 52% of children treated with mirdametinib had a confirmed radiographic response by blinded independent central review, defined as ≥ 20% volumetric reduction in target PN on ≥ 2 consecutive scans within 2–6 months [23]. Moreover, both selumetinib and mirdametinib showed improvements in patient-reported outcomes, including pain intensity and pain-interference [22, 23]. Other targeted therapies are not currently FDA approved for the treatment of NF1-PN but have been, or are currently being, assessed for efficacy and safety in clinical trials among adult [24–26], pediatric [27, 28], or both [29–32] populations of patients with NF1-PN. The disease burden associated with NF1-PN in adults is not well understood, and at the time this study was performed there were no FDA-approved treatments for NF1-PN in the adult patient population. To our knowledge, only one study has previously described healthcare resource utilization (HRU) and costs among adults with NF1-PN, evaluating a combined pediatric and adult NF1-PN population in Japan [33]. Thus, the objectives of this real-world study were to both characterize current treatment patterns and estimate all-cause and NF1/PN-specific HRU and costs among adult patients with NF1-PN in the US. Methods Data source De-identified administrative claims data were obtained from the Merative™ MarketScan® Commercial and Medicare Databases from January 1, 2016, to December 31, 2022. The MarketScan® databases contain claims from various healthcare encounter sources, including inpatient (IP), outpatient (OP), pharmacy, and long-term care services, for approximately 250 million unique individual employees, their spouses, and dependents covered by employer-sponsored private health insurance. MarketScan is a registered trademark of Merative Corporation in the US, other countries, or both. Available data comprises enrollment eligibility and demographic characteristics, as well as medical, surgical, and prescription drug utilization, and expenditure information. This study was considered exempt research under 45 CFR § 46.104(d)(4) as it involved only the secondary use of data that were de-identified in compliance with the Health Insurance Portability and Accountability Act (HIPAA), specifically, 45 CFR § 164.514. Additionally, it was conducted according to the standards of Good Pharmacoepidemiology Practice (GPP) and all applicable federal, state, and local laws, rules, and regulations. Study design and patient population In this retrospective cohort study, patients with NF1-PN (i.e., both a NF1 and a PN diagnosis) were matched to control patients without NF1-PN at a 1:5 ratio. For patients with NF1-PN, the index date was defined as the first instance of having medical claims for both NF1 and PN, with the latter claim for NF1 or PN determining the index date. Index dates for the matched control cohort were assigned based on the index date of the matched patient with NF1-PN. The baseline period comprised the 12 months prior to the index date, while the follow-up period spanned from after the index date to the earlier of the end of data availability or the end of patient enrollment, with a minimum required follow-up of 3 months to ensure adequate measurement of study outcomes. Patients were censored at the earliest of end of enrollment or the end of data availability (December 31, 2022). Patients in the NF1-PN cohort were identified as having ≥ 1 medical claim with a diagnostic code for NF1 (International Classification of Diseases, 10th Revision, Clinical Modification [ICD-10-CM]: Q85.01) and ≥ 1 medical claim with a diagnostic code for PN (ICD-10-CM: D36.10). NF-1 PN patients were required to be aged ≥ 18 years at the index date and have ≥ 12 months of continuous eligibility prior to the index date and ≥ 3 months of continuous eligibility after the index date. Patients who participated in a clinical trial during the baseline or follow-up period (ICD-10-CM: Z00.6; Healthcare Common Procedure Coding System [HCPCS]: S9988, S9990, S9991, S9992, S9994, S9996) were excluded. Patients who were covered by a capitated health plan were excluded from the cost analyses (Fig. 1). Sample selection flowchart. Abbreviations: HCPCS: healthcare common procedure coding system; HMO: health maintenance organization; ICD-10-CM: International Classification of Diseases, 10th Revision, Clinical Modification; NF1: neurofibromatosis type 1; PN: plexiform neurofibromas; POS: point-of-service. Note: [1] Among the final sample, 124 (13.1%) patients were covered by capitated health plans (i.e., HMO, and POS) and were excluded from cost analyses Patients without NF1-PN, defined as individuals with no recorded diagnosis of NF1 and no recorded diagnosis of PN at any time during the study period, were exact matched to patients with NF1-PN based on the following pre-defined sociodemographic characteristics: exact birth year, sex, region at index date, insurance type at index date, and COVID-19 period at index date (pre-pandemic: prior to March 15, 2020; COVID pre-vaccine: March 15, 2020 to March 8, 2021; COVID post-vaccine: March 9, 2021 or later). Controls were also required to have ≥ 12 months of continuous eligibility prior to the index date. Study outcomes Treatments were described for the NF1-PN cohort during the follow-up period and included debulking surgeries and procedures, prescription pain medications, targeted therapies [12, 34], cytotoxic chemotherapy, and radiotherapy (see Additional file 1 for list of codes). All-cause, NF1/PN-specific (i.e., HRU related to either the NF1 or PN diagnosis), and PN-specific HRU were assessed during the follow-up period for all patients and, separately, for patients with ≥ 1 type of visit (IP admission, emergency department [ED] visit, OP visit). All-cause HRU was described among the NF1-PN and matched control cohorts, separately, while NF1/PN-specific and PN-specific HRU were described among the NF1-PN cohort only. NF1/PN-specific HRU comprised medical claims with NF1 or PN as the primary or secondary diagnosis, while PN-specific HRU comprised medical claims for a specific encounter with PN as the primary or secondary diagnosis. Reported outcomes included HRU associated with IP admissions (proportion of patients with IP admissions, number of IP admissions, total length of stay), ED visits (proportion of patients with ED visits, number of ED visits), and OP visits (proportion of patients with OP visits, number of OP visits [including physical therapy and occupational therapy (PT/OT) visits, imaging procedures, specialist visits [35]; see Additional file 1 for list of codes]) . All-cause, NF1/PN-specific, and PN-specific healthcare costs were evaluated during the follow-up period, based on definitions equivalent to those applied to the HRU analysis. Reported outcomes included total healthcare costs, total medical costs (IP admission costs, ED costs, OP costs, other medical costs), and OP pharmacy costs. Statistical analysis Baseline demographic (age, sex, region, health plan type, COVID-19 period) and clinical (comorbidities, use of pain medications, HRU) characteristics were described among both the NF1-PN and matched control cohorts, separately, during the baseline period or on the index date. Descriptive statistics were reported as mean, standard deviation (SD), median, and interquartile range (IQR) for continuous variables, and frequency and proportion for categorical variables. Non-matching characteristics were compared between the NF1-PN and matched control cohorts using standardized mean differences (SMD), with an SMD > 10% considered an indicator of imbalance. Each treatment assessed during the follow-up period was reported as either the number and proportion of patients who received the treatment of interest, or the number of treatments per year using mean, SD, median, and IQR. For radiotherapy, treatment sessions and office visits that occurred on the same day were counted as a single visit; the same rule was applied to the calculation of surgery rates. All-cause HRU and costs were reported per patient per year (PPPY) for continuous variables, to account for varying lengths of follow-up, or described using frequency and proportions for categorical variables. PPPY estimates assume that observed HRU and costs during the available follow-up period are representative of the patient’s experience over time. To compare all-cause HRU between the NF1-PN and matched control cohorts, adjusted incidence rate ratios (IRRs) with their respective 95% confidence intervals (CIs) were estimated using a generalized linear model (GLM) with Poisson distribution. Matching variables, including birth year, sex, region, healthcare plan, and COVID-19 period at the index date, were included in the adjusted model for a doubly robust estimate. Additionally, the Quan-Charlson comorbidity index (Quan-CCI), which was unbalanced between the 2 cohorts and not considered an intermediate factor between NF1-PN and HRU, was adjusted in the regression model. To compare all-cause healthcare costs between the 2 cohorts, adjusted cost difference and cost ratios were estimated using GLM with Tweedie distribution using the same set of covariates as the HRU analysis. The corresponding 95% CIs and p-values were derived using bootstrapping and costs were adjusted for inflation using the Consumer Price Index (CPI) from the US Bureau of Labor Statistics to 2023 US dollars (USD). Results Baseline characteristics for patients with NF1-PN and matched controls A total of 944 patients with NF1-PN were identified (Fig. 1) and matched to 4,720 control patients without NF1-PN. Among patients with NF1-PN and their matched controls, the mean (SD) age was 39.6 (15.6) years, 59.6% were female, and 41.7% of patients were from the South US census region (Table 1). More than half had a preferred provider organization (PPO) coverage (54.7%), and most were identified during the pre-COVID-19 pandemic period (64.1%). A total of 124 (13.1%) patients with NF1-PN and their matched controls were enrolled in health maintenance organization (HMO) or capitated or partially capitated point-of-service (POS) plans and were excluded from the cost analyses. Table 1Baseline Demographics of Adult NF1-PN Patients and Matched Controls without NF1-PNPatient characteristicsNF1-PN cohort [A]Control cohorta [B]SMD (%)b [A] vs. [B]N = 944N = 4,720Patient demographicsAgec, years, mean ± SD [median] (IQR)39.6 ± 15.6 [39] (24, 53)39.6 ± 15.6 [39] (24, 53)Sexc, n (%) Female563 (59.6)2,815 (59.6)- Male381 (40.4)1,905 (40.4)-Geographic regionc, n (%) South394 (41.7)1,970 (41.7)- Northeast204 (21.6)1,020 (21.6)- North Central202 (21.4)1,010 (21.4)- West143 (15.1)715 (15.1)- Unknown1 (0.1)5 (0.1)-Insurance typec, d, n (%) PPO516 (54.7)2,580 (54.7)- HMO121 (12.8)605 (12.8)- CDHC99 (10.5)495 (10.5)- HDH95 (10.1)475 (10.1)- Non-capitated POS78 (8.3)390 (8.3)- Comprehensive18 (1.9)90 (1.9)- EPO7 (0.7)35 (0.7)- Capitated or partially capitated POS3 (0.3)15 (0.3)- Basic plan0 (0.0)0 (0.0)- Unknown7 (0.7)35 (0.7)-COVID-19 periods at index datec, n (%) Pre-pandemic (1/1/2016–3/14/2020)605 (64.1)3,025 (64.1)- COVID pre-vaccine (3/15/2020–3/8/2021)106 (11.2)530 (11.2)- COVID post-vaccine (3/9/2021–12/31/2022)233 (24.7)1,165 (24.7)-Clinical characteristicsCormobiditiese, f, n (%) Hypertension211 (22.4)841 (17.8)11.3%† Anxiety disorders187 (19.8)629 (13.3)17.5%† Abdominal pain144 (15.3)530 (11.2)11.9%† Depression/bipolar disorder137 (14.5)436 (9.2)16.4%† Insomnia80 (8.5)165 (3.5)21.1%† Attention deficit hyperactivity disorder65 (6.9)145 (3.1)17.6%† Constipation64 (6.8)97 (2.1)23.1%† Scoliosis58 (6.1)31 (0.7)30.6%† Headaches42 (4.4)93 (2.0)14.1%† Epilepsy/seizures36 (3.8)34 (0.7)20.9%† Neuropathy/peripheral neuropathy36 (3.8)21 (0.4)23.5%† Osteoporosis28 (3.0)49 (1.0)13.8%† Speech impairment24 (2.5)19 (0.4)17.8%† Congenital heart disease15 (1.6)15 (0.3)13.1%† Irritable bowel syndrome16 (1.7)67 (1.4)2.2% Benign brain tumors16 (1.7)2 (0.0)17.9%† Cognitive impairment11 (1.2)13 (0.3)10.5%† Aqueductal stenosis10 (1.1)1 (0.0)14.2%† Nonossifying fibromas9 (1.0)12 (0.3)9.0% Cancers164 (17.4)114 (2.4)51.7%† Melanoma49 (5.2)2 (0.0)32.7%† Soft-tissue sarcomag38 (4.0)2 (0.0)28.5%† MPNST29 (3.1)0.0 (0.0)25.2%† Breast cancer22 (2.3)44 (0.9)11.1%† Gastrointestinal stromal tumor13 (1.4)0.0 (0.0)16.7%†Quan-CCI scoree, mean ± SD [median] (IQR)0.83 ± 1.48 [0] (0, 2)0.25 ± 0.74 [0] (0, 0)50.2%†Use of prescription pain medicationse, n (%)532 (56.4)1,606 (34.0)46.0%†All-cause HRU during 12-month baseline period Had ≥ 1 IP admission, n (%)105 (11.1)200 (4.2)26.1%† Had ≥ 1 ED visit, n (%)350 (37.1)1,368 (29.0)17.3%† Had ≥ 1 OP visit, n (%)928 (98.3)4,082 (86.5)45.7%†Notes:aPatients with NF1-PN were matched without replacement to those without NF1-PN (1:5 ratio) on birth year, sex, region at index date, insurance type at index date, and COVID-19 periods at index date. Controls were required to have ≥ 12 months of continuous eligibility pre-indexbThe SMD was calculated as the difference in mean between 2 cohorts divided by the SD of the 2 cohorts pooled together. The absolute value of a SMD > 10% was considered as an indicator for imbalance between the cohorts. Co-morbidities with SMD > 20% were highlighted in the main text to emphasize the most pronounced differencescEvaluated at the index date (i.e., the date of first recorded medical claim for NF1 or PN, whichever occurs later, post 12 months of baseline). Age groups were based on US Census Bureau. https://www.census.gov/library/visualizations/interactive/exploring-age-groups-in-the-2020-census.htmldHMO plan, along with capitated or partially capitated POS plan, were considered capitated planseEvaluated during the 12-month baseline period, not including the index datefComorbidities observed among < 1% of patients were not shown, which included autism, interstitial lung disease, pulmonary hypertension and pulmonary artery stenosis, dyspepsia, learning disorders, pseudarthrosis, bone dysplasia, renal tubular and glomerular dysfunction, Chiari malformation, and vasculopathy (including Moyamoya disease). See Additional file 1 for a list of codes used to identify comorbiditiesgSoft-tissue sarcoma was defined by ICD-10 code C49.x. Although this should represent non-MPNST sarcomas, there may be some non-specific coding leading to overlap in the MPNST and soft-tissue sarcoma comorbidity groupsAbbreviations: AIDS acquired immune deficiency syndrome, CDHC consumer-directed healthcare, EPO exclusive provider organization, ED emergency department, HDH high-deductible health plan, HMO health maintenance organization, HRU healthcare resource utilization, ICD International Classification of Diseases, IP inpatient, IQR interquartile range, MPNST malignant peripheral nerve sheath tumor, NF1 neurofibromatosis type 1, OP outpatient, PN plexiform neurofibromas, POS point-of-service, PPO preferred provider organization, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation, SE standard error, SMD standardized mean differences, USD United States dollars† Represents SMD > 10% Baseline Demographics of Adult NF1-PN Patients and Matched Controls without NF1-PN Notes: aPatients with NF1-PN were matched without replacement to those without NF1-PN (1:5 ratio) on birth year, sex, region at index date, insurance type at index date, and COVID-19 periods at index date. Controls were required to have ≥ 12 months of continuous eligibility pre-index bThe SMD was calculated as the difference in mean between 2 cohorts divided by the SD of the 2 cohorts pooled together. The absolute value of a SMD > 10% was considered as an indicator for imbalance between the cohorts. Co-morbidities with SMD > 20% were highlighted in the main text to emphasize the most pronounced differences cEvaluated at the index date (i.e., the date of first recorded medical claim for NF1 or PN, whichever occurs later, post 12 months of baseline). Age groups were based on US Census Bureau. https://www.census.gov/library/visualizations/interactive/exploring-age-groups-in-the-2020-census.html dHMO plan, along with capitated or partially capitated POS plan, were considered capitated plans eEvaluated during the 12-month baseline period, not including the index date fComorbidities observed among < 1% of patients were not shown, which included autism, interstitial lung disease, pulmonary hypertension and pulmonary artery stenosis, dyspepsia, learning disorders, pseudarthrosis, bone dysplasia, renal tubular and glomerular dysfunction, Chiari malformation, and vasculopathy (including Moyamoya disease). See Additional file 1 for a list of codes used to identify comorbidities gSoft-tissue sarcoma was defined by ICD-10 code C49.x. Although this should represent non-MPNST sarcomas, there may be some non-specific coding leading to overlap in the MPNST and soft-tissue sarcoma comorbidity groups Abbreviations: AIDS acquired immune deficiency syndrome, CDHC consumer-directed healthcare, EPO exclusive provider organization, ED emergency department, HDH high-deductible health plan, HMO health maintenance organization, HRU healthcare resource utilization, ICD International Classification of Diseases, IP inpatient, IQR interquartile range, MPNST malignant peripheral nerve sheath tumor, NF1 neurofibromatosis type 1, OP outpatient, PN plexiform neurofibromas, POS point-of-service, PPO preferred provider organization, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation, SE standard error, SMD standardized mean differences, USD United States dollars † Represents SMD > 10% In the NF1-PN cohort, hypertension was the most frequent NF1-PN-related comorbidity (22.4%), followed by anxiety disorders (19.8%), cancers (17.4%; melanoma 5.2%, soft-tissue sarcoma 4.0%, malignant peripheral nerve sheath tumor [MPNST] 3.1%), abdominal pain (15.3%), and depression/bipolar disorder (14.5%). Differences in the prevalence of comorbidities between the NF1-PN and control cohorts are presented in Table 1. Comorbidities observed more often among the NF1-PN versus matched control cohort, with a SMD > 20%, included melanoma (5.2% vs. 0.0%), scoliosis (6.1% vs. 0.7%), soft-tissue sarcoma (4.0% vs. 0.0%), MPNST (3.1% vs. 0.0%), neuropathy/peripheral neuropathy (3.8% vs. 0.4%), constipation (6.8% vs. 2.1%), insomnia (8.5% vs. 3.5%), and epilepsy/seizures (3.8% vs. 0.7%). The mean Quan-CCI was significantly higher in the NF1-PN cohort compared to the matched control cohort (0.83 vs. 0.25, SMD: 50.2%). In total, 11.1% of patients with NF1-PN had ≥ 1 IP admission during the baseline period, while 37.1% had ≥ 1 ED visit, and nearly all (98.3%) had ≥ 1 OP visit. Among the matched control cohort, 4.2% had ≥ 1 IP admission, 29.0% had ≥ 1 ED visit, and 86.5% of patients had ≥ 1 OP visit. Treatment during the follow-up period Over a mean (SD) follow-up of 25.8 (19.7) months, 69.5% of patients with NF1-PN received prescription pain medication (Table 2). Other treatments among patients with NF1-PN included debulking surgeries (22.9%), cytotoxic chemotherapy (7.0%), and radiotherapy (4.8%). Targeted therapies (e.g., MEK inhibitors) were less commonly used (4.7%) during this time period prior to MEK inhibitor approval in the US for adults with NF1-PN. Among the 44 patients who received targeted therapies, 31 received MEK inhibitors and 11 received tyrosine kinase inhibitors (see Additional file 1 for list of targeted therapies and associated codes). Among the 216 patients who underwent ≥ 1 debulking surgery, most received only a single procedure during the entire follow-up period; consequently, the mean (SD) annualized rate was 0.55 (0.66) procedures per year over an average follow-up period of more than 2 years. Among the 45 patients with ≥ 1 record of radiotherapy, patients underwent a mean (SD) of 5.7 (9.5) sessions per year. The specific diagnosis for which treatments were given, including chemotherapy and radiation, was also unknown. Table 2Treatment among Adult Patients with NF1-PNTreatment patternsNF1-PN cohortN = 944Duration of follow-up period in months Mean ± SD [median] (IQR)25.8 ± 19.7 [20] (10, 37)Debulking surgeries and proceduresa Received ≥ 1 surgery/procedure, n (%)216 (22.9) Number of surgeries and proceduresb, PPPYc Mean ± SD [median] (IQR)0.55 ± 0.66 [0] (0, 1)Prescription pain medicationd Received ≥ 1 administration/dispensing, n (%)656 (69.5)Targeted therapiesa, n (%) Received ≥ 1 administration/dispensing44 (4.7) MEK inhibitors31 (3.3) Tyrosine kinase inhibitors11 (1.2) Pegylated interferon alpha-2b1 (0.1) Everolimus1 (0.1)Cytotoxic chemotherapy, n (%) Received ≥ 1 administration/dispensing, n (%)66 (7.0)Radiotherapye Received ≥ 1 radiotherapy, n (%)45 (4.8) Number of radiotherapyb, PPPY Mean ± SD [median] (IQR)5.7 ± 9.5 [2] (0, 11)Notes:aSee Additional file 1 for a list of procedure and drug codesbRates were calculated using data up to 12 months of follow-up and among patients who received at least one treatment within 12 months of follow-up. The observation period was truncated at 12 months. Rates were calculated as the sum of events divided by total duration of follow-up (in person-years). To calculate SD, the events were weighted by each patient’s length of follow-up to avoid overestimating the event by annualizing data for patients observed for less than 1 yearcAmong patients with at least one procedure, the PPPY rate may be less than 1 if the procedure occurred over a patient level follow-up period of longer than 1 yeardSee Additional file 1for a list of drug codeseSee Additional file 1 for a list of procedure and diagnosis codesAbbreviations: IQR interquartile range, MEK mitogen-activated and extracellular signal-regulated kinase, NF1 neurofibromatosis type 1, PN plexiform neurofibromas, PPPY per patient per year, SD standard deviation Treatment among Adult Patients with NF1-PN Notes: aSee Additional file 1 for a list of procedure and drug codes bRates were calculated using data up to 12 months of follow-up and among patients who received at least one treatment within 12 months of follow-up. The observation period was truncated at 12 months. Rates were calculated as the sum of events divided by total duration of follow-up (in person-years). To calculate SD, the events were weighted by each patient’s length of follow-up to avoid overestimating the event by annualizing data for patients observed for less than 1 year cAmong patients with at least one procedure, the PPPY rate may be less than 1 if the procedure occurred over a patient level follow-up period of longer than 1 year dSee Additional file 1for a list of drug codes eSee Additional file 1 for a list of procedure and diagnosis codes Abbreviations: IQR interquartile range, MEK mitogen-activated and extracellular signal-regulated kinase, NF1 neurofibromatosis type 1, PN plexiform neurofibromas, PPPY per patient per year, SD standard deviation All-cause HRU between patients with NF1-PN and matched controls In total, 20.9% of patients with NF1-PN had ≥ 1 IP admission. Among those with ≥ 1 IP admission, patients experienced a mean (SD) of 2.0 (1.9) IP admissions over the follow-up period, corresponding to an annualized mean (SD) of 0.81 (1.03) IP admissions PPPY and 6.9 (14.7) IP days PPPY over an average follow-up period of more than 2 years (Table 3). After adjusting for potential confounders, including birth year, sex, region, health care plan, COVID-19 period, and Quan-CCI score, patients with NF1-PN had a 3.22 times higher rate of IP admissions PPPY compared to patients without NF1-PN (95% CI: 2.78–3.72, p < 0.001). After adjustment, the mean number of IP days PPPY was 4.19 times higher in the NF1-PN versus matched control cohort (95% CI: 3.97–4.43, p < 0.001). Table 3All-Cause HRU among Adult Patients with NF1-PN and Matched ControlsHealthcare resource utilizationbNF1-PN cohort [A]Control cohort [B]Adjusted IRRa [A] / [B]N = 944N = 4,720IRR (95%CI)Duration of follow-up period in months, mean ± SD [median] (IQR)25.8 ± 19.7 [20] (10, 37)24.5 ± 19.8 [19] (9, 36)IP admission Number of admissions, PPPY, mean ± SD0.19 ± 0.610.05 ± 0.193.22 (2.78–3.72)* Total IP days, PPPY, mean ± SD1.6 ± 7.70.27 ± 3.284.19 (3.97–4.43)* Had ≥ 1 IP admission, n (%)197 (20.9)336 (7.1) Number of admissions, PPPYc, mean ± SD0.81 ± 1.030.45 ± 0.42- Total IP days, PPPYc, mean ± SD6.9 ± 14.72.7 ± 9.9- Number of admissions, per patient, mean ± SD2.0 ± 1.91.3 ± 0.7 Total IP days, per patient, mean ± SD16.7 ± 27.57.8 ± 16.2ED visit Number of ED visits, PPPY, mean ± SD0.80 ± 1.760.53 ± 0.981.36 (1.28–1.44)* Had ≥ 1 ED visit, n (%)482 (51.1)1,933 (41.0) Number of visits, PPPYc, mean ± SD1.3 ± 2.10.95 ± 1.15-OP visit Number of OP visits, PPPY, mean ± SD18.8 ± 18.29.1 ± 11.61.81 (1.78–1.83)* Number of PT/OT visitsd, PPPY, mean ± SD2.5 ± 6.41.0 ± 3.42.17 (2.09–2.25)* Number of imaging procedurese, PPPY, mean ± SD2.4 ± 2.80.88 ± 1.322.45 (2.36–2.54)* Number of specialist visitsf, PPPY, mean ± SD2.0 ± 2.50.47 ± 2.003.64 (3.48–3.81)* Had ≥ 1 OP visit, n (%)943 (99.9)4,047 (85.7) Number of visits, PPPY, mean ± SD18.8 ± 18.29.6 ± 11.7- Had ≥ 1 PT/OT visitd, n (%)295 (31.3)816 (17.3) Number of visits, PPPY, mean ± SD21.6 ± 32.84.2 ± 5.8- Had ≥ 1 imaging procedure visite, n (%)768 (81.4)2,247 (47.6) Number of procedures, PPPY, mean ± SD2.7 ± 2.81.4 ± 1.4- Had ≥ 1 specialist visitf, n (%)714 (75.6)1,132 (24.0) Number of visits, PPPY, mean ± SD2.4 ± 2.61.5 ± 3.3-Notes:aIRR was estimated using generalized linear model with Poisson distribution. Matching variables including birth year, sex, region, healthcare plan, and COVID-19 period at the index date, plus Quan-CCI score were included in the adjusted model for a doubly robust estimatebTotal IP days and mean HRU numbers (i.e., IP admission, ER visits, OP visits) were calculated across all patients, regardless of whether a patient had incurred HRU, unless specified. Patients without any HRU were given a value of 0cAmong patients with at least one observed IP admission or ED visit, the PPPY rate may be less than 1 if the event occurred over a patient level follow-up period of longer than 1 yeardPT/OT visits were identified using procedure codes and revenue codes. See Additional file 1 for detailseSee Additional file 1 for a list of procedure and diagnosis codes related to imagingfSpecialist visit was identified as OP visit with stdprov codes: 140, 215, 260, 325, 355, 380, 520. See Additional file 1 for detailsAbbreviations: CI confidence interval, ED emergency department, IP inpatient, IRR incidence rate ratio, IQR interquartile range, NF1 neurofibromatosis type 1, OP outpatient, OT occupational therapy, PN plexiform neurofibromas, PPPY per patient per year, PT physical therapy, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation* Statistically significant difference with p value < 0.001 All-Cause HRU among Adult Patients with NF1-PN and Matched Controls Notes: aIRR was estimated using generalized linear model with Poisson distribution. Matching variables including birth year, sex, region, healthcare plan, and COVID-19 period at the index date, plus Quan-CCI score were included in the adjusted model for a doubly robust estimate bTotal IP days and mean HRU numbers (i.e., IP admission, ER visits, OP visits) were calculated across all patients, regardless of whether a patient had incurred HRU, unless specified. Patients without any HRU were given a value of 0 cAmong patients with at least one observed IP admission or ED visit, the PPPY rate may be less than 1 if the event occurred over a patient level follow-up period of longer than 1 year dPT/OT visits were identified using procedure codes and revenue codes. See Additional file 1 for details eSee Additional file 1 for a list of procedure and diagnosis codes related to imaging fSpecialist visit was identified as OP visit with stdprov codes: 140, 215, 260, 325, 355, 380, 520. See Additional file 1 for details Abbreviations: CI confidence interval, ED emergency department, IP inpatient, IRR incidence rate ratio, IQR interquartile range, NF1 neurofibromatosis type 1, OP outpatient, OT occupational therapy, PN plexiform neurofibromas, PPPY per patient per year, PT physical therapy, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation * Statistically significant difference with p value < 0.001 Among the NF1-PN cohort, 51.1% of patients had ≥ 1 ED visit, with a mean (SD) of 1.3 (2.1) ED visits PPPY among those with ≥ 1 ED visit. The adjusted rate of ED visits PPPY was 1.36 times higher in patients with NF1-PN compared to matched controls (95% CI: 1.28–1.44, p < 0.001). Almost all patients (99.9%) in the NF1-PN cohort had ≥ 1 OP visit, with a mean (SD) of 18.8 (18.2) OP visits PPPY among those with ≥ 1 OP visit. The adjusted rate of OP visits PPPY was 1.81 higher among the NF1-PN versus matched control cohort (95% CI: 1.78–1.83, p < 0.001). In total, 31.1%, 81.4% and 75.6% of patients with NF1-PN had ≥ 1 PT/OT visit, ≥ 1 imaging procedure visit, or ≥ 1 specialist visit during follow-up, respectively. The adjusted rate of PT/OT visits PPPY (2.17 [95% CI: 2.09–2.25]), imaging procedures PPPY (2.45 [95% CI: 2.36–2.54]), and specialist visits PPPY (3.64 [95% CI: 3.48–3.81]) were all significantly higher among the NF1-PN versus the matched control cohort (all p < 0.001). NF1/PN- and PN-specific HRU among patients with NF1-PN Out of a mean (SD) of 1.6 (7.7) all-cause IP days PPPY, 0.84 (5.96) IP days PPPY (52.5%) were attributable to an NF1 or PN diagnosis, defined as NF1 or PN being noted as the primary or secondary diagnosis, and 0.45 (4.67) IP days PPPY (28.1%) were attributable to a PN diagnosis (Table 4). Out of a mean (SD) of 18.8 (18.2) all-cause OP visits PPPY, 3.3 (5.2) OP visits PPPY (17.6%) were attributable to an NF1 or PN diagnosis and 1.2 (1.8) OP visits PPPY (6.4%) were attributable to a PN diagnosis. Patients with NF1-PN had a mean (SD) of 2.5 (6.4) PT/OT visits PPPY, 2.4 (2.8) imaging procedure visits PPPY, and 2.0 (2.5) specialist visits PPPY. In total, a mean (SD) of 0.26 (1.61) PT/OT visits PPPY (10.4%), 0.67 (1.11) imaging procedure visits PPPY (27.9%), and 0.88 (1.20) specialist visits PPPY (44.0%) were attributable to an NF1 or PN diagnosis, while 0.03 (0.24) PT/OT visits PPPY (1.2%), 0.26 (0.60) imaging procedure visits PPPY (10.8%), and 0.41 (0.74) specialist visits PPPY (20.5%) were attributable to a PN diagnosis. Table 4NF1/PN- and PN-specific HRU among Adult Patients with NF1-PNHealthcare resource utilizationaAll-cause HRUNF1/PN-specific HRUbPN-specific HRUbN = 944N = 944N = 944Duration of follow-up period in months, mean ± SD [median] (IQR)25.8 ± 19.7 [20] (10, 37)24.5 ± 19.8 [19] (9, 36)IP admission Number of admissions, PPPY, mean ± SD0.19 ± 0.610.09 ± 0.390.04 ± 0.19 Total IP days, PPPY, mean ± SD1.6 ± 7.70.84 ± 5.960.45 ± 4.67 Had ≥ 1 IP admission, n (%)197 (20.9)112 (11.9)67 (7.1) Number of admissions, PPPYc, mean ± SD0.81 ± 1.030.61 ± 0.860.52 ± 0.46 Total IP days, PPPYc, mean ± SD6.9 ± 14.76.0 ± 15.05.7 ± 15.7 Number of admissions, per patient, mean ± SD2.0 ± 1.91.6 ± 1.41.2 ± 0.6 Total IP days, per patient, mean ± SD16.7 ± 27.515.1 ± 25.113.5 ± 20.0ED visit Number of ED visits, PPPY, mean ± SD0.80 ± 1.760.07 ± 0.470.02 ± 0.13 Had ≥ 1 ED visit, n (%)482 (51.1)79 (8.4)29 (3.1) Number of visits, PPPYc, mean ± SD1.3 ± 2.10.68 ± 1.320.63 ± 0.51OP visit Number of OP visits, PPPY, mean ± SD18.8 ± 18.23.3 ± 5.21.2 ± 1.8 Number of PT/OT visitsd, PPPY, mean ± SD2.5 ± 6.40.26 ± 1.610.03 ± 0.24 Number of imaging procedurese, PPPY, mean ± SD2.4 ± 2.80.67 ± 1.110.26 ± 0.60 Number of specialist visitsf, PPPY, mean ± SD2.0 ± 2.50.88 ± 1.200.41 ± 0.74 Had ≥ 1 OP visit, n (%)943 (99.9)924 (97.9)801 (84.9) Number of visits, PPPY, mean ± SD18.8 ± 18.23.4 ± 5.21.3 ± 1.9 Had ≥ 1 PT/OT visitd, n (%)295 (31.3)74 (7.8)31 (3.3) Number of visits, PPPY, mean ± SD21.6 ± 32.89.6 ± 17.42.8 ± 3.4 Had ≥ 1 imaging procedure visite, n (%)768 (81.4)489 (51.8)300 (31.8) Number of procedures, PPPY, mean ± SD2.7 ± 2.81.1 ± 1.22.0 ± 2.0 Had ≥ 1 specialist visitf, n (%)714 (75.6)590 (62.5)429 (45.4) Number of visits, PPPY, mean ± SD2.4 ± 2.61.3 ± 1.32.4 ± 2.4Notes:aTotal IP days and mean HRU numbers (i.e., IP admission, ER visits, OP visits) were calculated across all patients, regardless of whether a patient had incurred HRU, unless specified. Patients without any HRU were given a value of 0bNF1/PN-specific HRU comprised medical claims with NF1 or PN as the primary or secondary diagnosis, while PN-specific HRU comprised medical claims for a specific encounter with PN as the primary or secondary diagnosiscAmong patients with at least one observed IP admission or ED visit, the PPPY rate may be less than 1 if the event occurred over a patient level follow-up period of longer than 1 yeardPT/OT visits were identified using procedure codes and revenue codes. See Additional file 1 for detailseSee Additional file 1 for a list of procedure and diagnosis codes related to imagingfSpecialist visit was identified as OP visit with stdprov codes: 140, 215, 260, 325, 355, 380, 520. See Additional file 1 for detailsAbbreviations: CI confidence interval, ED emergency department, IP inpatient, IRR incidence rate ratio, IQR interquartile range, NF1 neurofibromatosis type 1, OP outpatient, OT occupational therapy, PN plexiform neurofibromas, PPPY per patient per year, PT physical therapy, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation NF1/PN- and PN-specific HRU among Adult Patients with NF1-PN Notes: aTotal IP days and mean HRU numbers (i.e., IP admission, ER visits, OP visits) were calculated across all patients, regardless of whether a patient had incurred HRU, unless specified. Patients without any HRU were given a value of 0 bNF1/PN-specific HRU comprised medical claims with NF1 or PN as the primary or secondary diagnosis, while PN-specific HRU comprised medical claims for a specific encounter with PN as the primary or secondary diagnosis cAmong patients with at least one observed IP admission or ED visit, the PPPY rate may be less than 1 if the event occurred over a patient level follow-up period of longer than 1 year dPT/OT visits were identified using procedure codes and revenue codes. See Additional file 1 for details eSee Additional file 1 for a list of procedure and diagnosis codes related to imaging fSpecialist visit was identified as OP visit with stdprov codes: 140, 215, 260, 325, 355, 380, 520. See Additional file 1 for details Abbreviations: CI confidence interval, ED emergency department, IP inpatient, IRR incidence rate ratio, IQR interquartile range, NF1 neurofibromatosis type 1, OP outpatient, OT occupational therapy, PN plexiform neurofibromas, PPPY per patient per year, PT physical therapy, Quan-CCI Quan-Charlson comorbidity index, SD standard deviation Healthcare costs During the follow-up period, the mean all-cause total healthcare costs were $34,398 PPPY in the NF1-PN cohort versus $6,149 PPPY in the matched control cohort, with an adjusted cost difference of $23,516 (p < 0.001) (Fig. 2A). The adjusted cost ratio indicated that the mean total healthcare costs PPPY were 4.28 times higher for patients with NF1-PN versus those without (95% CI: 3.51–5.24, p < 0.001). The mean total healthcare costs in the NF1-PN cohort were largely driven by OP costs ($13,519 PPPY), followed by IP admission costs ($10,808 PPPY) and pharmacy costs ($8,738 PPPY). Total healthcare costs across all settings were significantly higher in the NF1-PN cohort compared to the matched control cohort, with the largest adjusted differences PPPY observed for OP visits (cost difference: $9,802; cost ratio: 4.13, 95% CI: 3.53–4.86, p < 0.001), followed by IP admissions (cost difference: $7,247; cost ratio: 4.81, 95% CI: 3.28–8.03, p < 0.001) and pharmacy costs (cost difference: $6,797; cost ratio: 4.98, 95% CI: 3.14–6.86, p < 0.001). Healthcare costs among patients with NF1-PN and matched controls. A All-cause costs and adjusted cost differences and cost ratios. B All-cause, NF1/PN-specific, and PN-specific costs among patients with NF1-PN1. Abbreviations: ∆ cost: adjusted cost difference derived from generalized linear model with Tweedie distribution, controlling for birth year, sex, region, healthcare plan, and COVID-19 period at the index date.; CR: adjusted cost ratio; ED: emergency department; IP: inpatient; NF1: neurofibromatosis; OP: outpatient; PN: plexiform neurofibromas. Note: [1] NF1/PN-specific costs comprised medical claims with NF1 or PN as the primary or secondary diagnosis, while PN-specific costs comprised medical claims for a specific encounter with PN as the primary or secondary diagnosis During the follow-up period, the mean all-cause total medical costs among patients with NF1-PN were $25,660 PPPY, with $11,109 PPPY (43.3%) attributable to an NF1 or PN diagnosis, and $5,638 PPPY (22.0%) attributable to a PN diagnosis (Fig. 2B). The mean all-cause IP admission costs were $10,808 PPPY, with $5,874 PPPY (54.3%) attributable to an NF1 or PN diagnosis and $3,335 PPPY (30.9%) attributable to a PN diagnosis. Similarly, the mean all-cause OP costs were $13,519 PPPY, with $5,026 (37.2%) attributable to an NF1 or PN diagnosis and $2,236 (16.5%) attributable to a PN diagnosis. Discussion This large population-based study comprehensively characterized current treatment patterns and estimated both all-cause and NF1/PN-specific HRU and costs across all encounter settings among adult patients with NF1-PN in the US. The most frequently administered treatment among patients with NF1-PN was prescription pain medication (70%), followed by debulking surgeries (23%), cytotoxic chemotherapy (7%), radiotherapy (5%), and targeted therapies (e.g., MEK inhibitors) (5%), indicating substantial needs related to treatment of ongoing symptoms among this patient population. All-cause HRU was consistently and significantly higher among patients with NF1-PN versus those without across all encounter settings (1.6 vs. 0.27 IP days PPPY; 18.8 vs. 9.6 OP visits PPPY). After adjustment for matching variables and comorbidity burden, patients with NF1-PN experienced 1.4 to 4.2 times higher rates of HRU across settings (all p < 0.001). Further, adjusted healthcare costs were $23,516 higher among patients with NF1-PN versus those without ($34,398 vs. $6,149 PPPY), with a cost ratio of 4.3 (p < 0.001), highlighting the significant economic burden associated with NF1-PN. Patients with NF1-PN often have extensive comorbidities. In the present study, conditions including scoliosis (6.1%), melanoma (5.2%), soft-tissue sarcoma (4.0%), epilepsy/seizures (3.8%), neuropathy/peripheral neuropathy (3.8%), and MPNST (3.1%) were observed among patients with NF1-PN while being rare among the matched control cohort (< 1.0%). These findings are consistent with the results of an earlier real-world study, where the most frequent comorbidities among a combined cohort of pediatric and adult patients with NF1-PN in Japan were epilepsy (8.7%), scoliosis (8.4%), and MPNST (3.8%) [33]. Soft-tissue sarcomas were also frequent, though this may reflect some non-specific coding practices with potential overlap between the sarcomas captured in the MPNST and other soft-tissue sarcoma categories. The most common treatment among patients with NF1-PN in the current analysis was prescription pain medication, received by 69.5% of patients, highlighting the substantial burden of pain-related symptoms among this population. Previous real-world claims-based studies have also identified high rates of pain medication use among both pediatric and adult patients with NF1-PN, ranging from 44.2% to 58.5% [33, 36, 37] Cytotoxic chemotherapy was observed among 7.0% of patients. As cytotoxic chemotherapy is not an approved treatment for NF1-PN and there is little evidence supporting its efficacy among this population, its use is typically reserved for highly selected patients, such as those requiring treatment for comorbid cancer [2, 38] Thus, its use here likely reflects the increased cancer risk associated with NF1, though a limitation of our dataset is that the diagnosis for which a treatment is given is not specified. A similar caveat applies to the observed rate of radiotherapy (4.8%) as this is rarely used in the treatment of NF1-PN and likely reflects treatment of comorbid malignancies or particularly refractory PNs [11]. Further, the number of radiotherapy sessions reported corresponds to the number of billable events (i.e., claims) versus the clinical fractions or complete radiation courses administered. As no targeted therapies were approved for the adult NF1-PN population at the time of our study conduct, the low prevalence of targeted therapy use (4.7%) was not unexpected. The proportion of patients administered targeted therapies was slightly higher than that previously reported among a Medicaid- (1.6%) or commercially- (1.0%) insured pediatric NF1-PN population prior to selumetinib approval [36, 37] Notably, the introduction of MEK inhibitors, including selumetinib and mirdametinib, represents an important advancement in the treatment of NF1-PN and has the potential to substantially alter treatment patterns as well as HRU and costs. While the use of targeted therapies may be associated with higher pharmacy costs, reductions in other components of HRU, including surgical interventions or symptom-related care, are probable [34, 39]. Accordingly, further real-world research is warranted to evaluate the overall impact of MEK inhibitors on NF1-PN management, HRU, and costs. This study is the first to quantify and compare the incremental healthcare and economic burden between adult patients with NF1-PN and matched controls in the US. Adjusted all-cause HRU rates were significantly higher among patients with NF1-PN versus those without across IP, ED, and OP settings, ranging from 1.36 to 4.19 fold adjusted differences in rates (all p < 0.001). Patients with NF1-PN had a significantly greater number of total IP days compared to patients without NF1-PN, and the proportion of patients with ≥ 1 IP admission was nearly 3 times higher among the NF1-PN versus the matched control cohort. Together, this may reflect the complexity of clinical management and severity of complications among the adult NF1-PN population. Although a very high proportion of patients both with and without NF1-PN had OP visits, approximately 2 times the proportion of patients among the NF1-PN cohort required ≥ 1 PT/OT, imaging, or specialist visit. Further, the adjusted rate of specialist visits was over 3 times higher among patients with NF1-PN versus those without, highlighting the specific needs of these patients. The absolute burden of HRU was similar to previous research among pediatric patients with NF1-PN in the US, with similar proportions of patients requiring IP visits (~ 13% versus 20.9%) and OP visits (~ 100% versus 99.9%) [36, 37]. The mean number of visits among pediatric patients with NF1-PN requiring healthcare services was also comparable to the adult NF1-PN population in the present study (IP visits: 1.2 to 1.4 PPPY versus 0.8 PPPY; ED visits: 1.3 to 1.6 PPPY versus 1.3 PPPY; OP visits: 17.3 to 18.2 PPPY versus 18.8 PPPY) [36, 37]. Only one other study has previously described HRU among adult patients with NF1-PN, evaluating a combined cohort of pediatric and adult patients in Japan [33]. Over a median follow-up of 33.6 months, 43.8% of patients with NF1-PN required ≥ 1 IP admission and nearly all (99.5%) required an OP visit. A total of 61.6% and 43.2% of patients with NF1-PN required magnetic resonance imaging or computed tomography imaging services, respectively. Although HRU rates among Japanese patients without NF1-PN was not measured, the proportion of patients with NF1-PN who accessed healthcare services are in line with findings of the present study and support the substantial economic burden associated with NF1-PN in adult patients across encounter settings. Among patients with NF1-PN, the proportion of HRU that was directly attributable to an NF1 or PN diagnosis or a PN diagnosis varied across settings. A descriptively higher proportion was attributable to the NF1 or PN diagnosis in the IP setting compared to the OP setting. The proportion of PT/OT visits attributable to an NF1 or PN diagnosis or a PN diagnosis was notably lower compared to imaging and specialist visits, potentially reflecting limited recognition of NF1-PN in driving the need for PT/OT care. It is also possible that the lower proportion of OP versus IP visits attributable to an NF1 and/or PN diagnosis could be due to differences in coding practice to capture the underlying reasons for medical visits in the OP setting. Consistent with the HRU analyses, all-cause healthcare costs were significantly higher among patients with NF1-PN versus those without across all settings. The mean total all-cause healthcare costs observed among adults with NF1-PN ($34,398 PPPY) were comparable to those previously identified among pediatric patients with NF1-PN covered by commercial insurance in the US ($38,292 PPPY) [36] and descriptively higher than for pediatric patients with NF1-PN insured with Medicaid ($17,275 PPPY) [37]. Taken together, the findings of the present study highlight the economic burden among adults with NF1-PN in the US and reflect the continued unmet medical need and the lack of approved treatments among this population. This study was subject to several limitations. As with all studies of large claims databases, this analysis was vulnerable to administrative coding inaccuracies, including miscoding and use of rule-out diagnoses, leading to potential misclassification of exposure groups and NF1-PN related outcomes. The study population may be skewed toward patients with more severe NF1-PN, as those with milder disease may not have had relevant ICD-10-CM codes documented in their claims. Treatments were identified based on pharmacy and procedure data, without knowing the specific diagnostic reasoning. Because claims databases only record diagnostic and procedure codes for reimbursement purposes, HRU and cost estimates may be either overestimated or underestimated due to coding variability and billing practices [40]. However, there is no reason to consider that this would have differentially impacted the NF1-PN and matched control cohorts. Clinical measures, such as the number, size, and operability of PNs, as well as the indication for treatment, were not available, limiting the interpretation of disease burden and whether this differentially affects certain patient subgroups. As this was a non-randomized observational study, it is possible that unmeasured confounding was not fully addressed despite matching between NF1-PN and control cohorts. In addition, although patients were matched on geographic region at index date, regional variation in treatment patterns and healthcare costs were not examined. Differential access to specialized NF1 care centers across regions may therefore have contributed to unmeasured heterogeneity in HRU and costs. The study period also spanned the COVID-19 pandemic, which may have influenced healthcare delivery, HRU, and costs over time. While matching on COVID-19 period was used to mitigate potential confounding, year-to-year variations in treatment patterns and expenditures were not separately evaluated. In addition, PPPY estimates assume that observed HRU during the available follow-up period is representative of the patient’s experience over time, though it is possible clinical trajectories may change. A fixed 12-month follow-up duration was not imposed for either cohort, to minimize selection bias related to differential enrollment or attrition patterns. Similarly, a minimum follow-up duration was not imposed for patients without NF1-PN; however, the observed follow-up duration was similar between the NF1-PN and control cohorts. Finally, the generalizability of results is constrained, as findings from this commercially insured population may not be fully applicable to the broader US population of patients with NF1-PN without insurance coverage. However, given the paucity of previous burden of disease data on adults with NF1-PN this study still represents a larger, more diverse population than previously studied. Conclusions In this real-world study, HRU and healthcare costs were consistently higher among adults with NF1-PN compared to those without, evidenced across all encounter settings. With no specific approved pharmacological treatments during the study period, these findings reinforce both the substantial economic burden associated with NF1-PN and the need for new treatments and interventions to effectively manage NF1-PN in the adult population. Supplementary Information Additional file 1: Lists of codes (XLSX). Acknowledgements Medical writing assistance was provided by professional medical writer, Molly Gingrich, MSc, an employee of Analysis Group, Inc., a consulting company that has provided paid consulting services to Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, which funded the development and conduct of this study and manuscript. Previous presentations Part of the material in this manuscript was presented at the 2025 Academy of Managed Care Pharmacy (AMCP) Annual Meeting held March 31 – April 3, 2025, Houston, TX as a poster presentation. Abbreviations Confidence interval Consumer Price Index Emergency department Food and Drug Administration Generalized linear model Good Pharmacoepidemiology Practice Healthcare Common Procedure Coding System Health Insurance Portability and Accountability Act Health maintenance organization Healthcare resource utilization International Classification of Diseases, 10th Revision, Clinical Modification Inpatient Interquartile range Incidence rate ratios Mitogen-activated protein kinase/extracellular signal-regulated kinase Mitogen-activated protein kinase Malignant peripheral nerve sheath tumor Neurofibromatosis type 1 Outpatient Objective response rate Occupational therapy Plexiform neurofibromas Point-of-service Preferred provider organization Per patient per year Physical therapy Quan-Charlson comorbidity index Standard deviation Standardized mean differences United States United States dollars Authors’ contributions All authors have made substantial contributions to the conception or design of the study, or the acquisition, analysis, or interpretation of data, drafting the manuscript and revising it critically for important intellectual content, and have provided final approval of this version to be published and agree to be accountable for all aspects of the work. XY, MO, TD, AA, JTJ, MC, QL, and MSD contributed to study conception and design, and result interpretation. XY, MO, MC, QL, DL, GC, and MSD contributed to collection and assembly of data and data analysis. All authors reviewed and approved the final content of this manuscript. Funding This study was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA. The study sponsor was involved in several aspects of the research, including the study design, interpretation of data, and preparation of the manuscript. Data availability The data that support the findings of this study are available from Merative™ MarketScan®. Restrictions apply to the availability of these data, which were used under license for this study. Declarations Ethics approval and consent to participate The study was considered exempt research under 45 CFR § 46.104(d)(4) as it involved only the secondary use of data that were de-identified in compliance with the Health Insurance Portability and Accountability Act (HIPAA), specifically, 45 CFR § 164.514. The study was conducted according to the standards of Good Pharmacoepidemiology Practice (GPP) and all applicable federal, state, and local laws, rules, and regulations. Consent for publication Not applicable. Competing interests XY and MO are employees of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA and own stock/stock options in Merck & Co., Inc., Rahway, NJ, USA. TD and AA are employees of Alexion, AstraZeneca Rare Disease and own stock/stock options. JTJ is an independent consultant of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA., Akeila Bio, Alexion Pharmaceuticals, AstraZeneca, Navio Theragnostics, Recursion Pharmaceuticals, Shepherd Therapeutics, Springworks Therapeutics, is an employee of University of Texas Southwestern Medical Center, and owns stock/stock options from Akeila Bio, Khora TX, Navio Theragnostics, and Shepherd Therapeutics. MC, QL, DL, GC, and MSD are employees of Analysis Group, Inc., a consulting company that has provided paid consulting services to Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, which funded the development and conduct of this study and manuscript. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Additional file 1: Lists of codes (XLSX). Data Availability Statement The data that support the findings of this study are available from Merative™ MarketScan®. Restrictions apply to the availability of these data, which were used under license for this study.